Preterm infants diagnosed with periventricular leukomalacia (PVL) are at high risk for cerebral palsy (CP), developmental delay and alterations in behavioral responsiveness. Infants who received the proposed intervention in the pilot study had almost 50% less ankle clonus, the best predictor of later CP, and an average hospital stay of 8 days less than controls. This research will evaluate selected physiological, behavioral and developmental responses of preterm infants with PVL to multi-sensory stimulation in the NICU and at home through two months corrected age. The specific aims are: 1) evaluate preterm infants' immediate physiological and behavioral responses to a multimodal sensory intervention; 2) quantify experimental and control group infants' neurobehavioral responses using multiple Brazelton assessments and neurodevelopmental protocols during hospitalization; 3) describe and compare experimental and control group infants' neuromotor and mental development during the first year of life and 4) Compare interactional behaviors of mothers and infants in the experimental and control groups. One hundred preterm infants with documented PVL and an additional 20-25 cavitary PVL infants will be randomly assigned to a control or experimental group at 33 weeks post conceptional age. Controls will receive no additional stimulation other than the Individualized Developmental Care and Assessment Program (NIDCAP) which will be in place for all infants in the MICU. Experimental infants will receive 15 minutes of auditory (female voice), tactile (massage), visual (eye to eye contact), and vestibular (rocking) stimulation twice a day, 5 days a week for 4 weeks during hospitalization and in the home through 2 months corrected age. All infants will have dependent measures recorded at the same time intervals. Immediate outcome measures include: heart and respiratory rate; arterial oxygen saturation; transcutaneous oxygen and carbon dioxide pressure; body temperature; and infant state. A Brazelton Neonatal Assessment and a neurodevelopmental examination will be conducted when the infant is 37-38 and 39-40 weeks post conceptional age. Mother-infant interaction will be recorded on videotape at these same time intervals and at 2 months corrected age. After hospital discharge, a Bayley-neurodevelopmental examination will be conducted a 2,4,8, & 12 months corrected age. Analyses will include chi-squares, z scores, repeated measures ANOVAS, and cluster analyses. If experimental infants demonstrate significant improvement in mental and/or neuromotor function, the intervention may be recommended to reduce developmental delay for PVL preterm infants.